200832162 九、發明說明： 【發明所屬之技術領域】 本發明是關於-種股市新聞分纖索系、统及方法，尤 指-稼藉由建立新聞與股票名稱之間的索引關聯，來縮短 檢鋒聞的時間、降低使用者端的運算負擔、以及減少資 料傳輸頻S：的-槪市新聞分類檢索系統及方法考。 【先前技術】 隨著資訊科技的突飛猛進，人們對於即時且方便的資 訊取得與通訊需求更是呈現爆炸性的成長。而各糗多樣化 功此之可攜式無線通訊裝置，也為了因應人們生活上的方 便與需要，而不斷地被開發出來。例如，各式手機（或稱 行動電話，Cellular Phone)、智慧型手機（Smartph〇ne)、 票及^具行動通訊功能之個人數位助理（Personal faigitalAssistant;簡稱PDA)等等。 除了别述支各種無線通訊裝置的硬體外，更有許多可 配合前述硬體來執行與應用的軟體與功能不斷地被開發 出來，以讓使用者可以更便利、更即時、且更隨時隨地的 理財、工作、娛樂或擷取資訊等等。例如，透過無線通訊 裝置來進行股票或期貨等有價證券的電子交易便是其中 一例。這些用來進行股票或期貨等有價證券之電子交易的 無線通訊裝置，除了進行電子下單之功能外，即時性之新 聞資訊的檢索與下載功能更是另一項不可或缺之重要功 能，以便能協助使用者可以更準確地去判斷這些有價證券 200832162 之償格波動趨勢，以及掌握最佳之買賣時機點。 請參閲圖一，其係為習知股市新聞檢索系統的示意 圖。目前習知的股市新聞檢索系統10,在自新聞來源= 端收到新聞後，一方面會把新聞儲存在一新聞資料庫1〇1 中，另一方面則會將該新聞單向的推送（Push)到與其連 線的無線通訊裝置（也就是使用者端12)，而完全無考慮 到使用者的實際需求。這些新聞資料，無論是否為使用者 所感興趣，都會被儲存在使用者端12中。使用者必須自 行再g這些新聞資料中一則一則去尋找與過濾出想閱讀 的新聞，不僅無法即時提供使用者真正需要的新聞，更浪 費了無蜂通訊有限的傳輸頻寬，也大量佔擄了使用者端 12的記憶體儲存容量。並且，在現有的新聞檢索技術中， 絕大部分都是將檢索模組121設置在使用者端12 (無線 通訊裝置)中，而增加了使用者進行新聞檢索時所需的時 間以及使用者端12的運算負擔者。 【發明内容】 本發明的主要目的是提供一種股市新聞分類檢索系 統及方法，其藉由建立新聞與股票名稱之間的索引關聯、 以及將檢索模組設置在系統端，來縮短檢索新聞的時間、 降低使用者端的運算負擔、以及減少資料傳輸頻寬的佔 用0 本發明的另一目的是提供一種股市新聞分類檢索系 統及方法，其可依據使用者端的請求，來過濾出使用者真 200832162 正感興趣的新聞，且只有這些使用者有興趣的新聞才會被 傳送至使用者端，以節省無線通訊頻寬與使用者端記憶艘 的使用量。 為達上述目的，本發明所提供之股市新聞分類檢索系 統係包括有··一股名分類資料庫、一索引關聯模組及一新 聞索引資料庫。股名分類資料庫中儲存有複數筆股名分 類。各股名分類分別更包括有一股票名稱、一股票代碼、 及一股票分類。索引關聯模組係用以分析新聞中所包含與 股票相關之關鍵字，並從該股名分類資料庫中找出與該關 鍵字相對應之一股名分類，並據以建立該新聞與該股名分 類之索引關聯。之後，再將該新聞與其所對應之索引關聯 儲存於新聞索引資料庫中。當使用者端欲檢索某新聞時， 僅需傳送特定關鍵字或代碼至該系統，該系統便能自新聞 索引資料庫中找出相符合之新聞並傳送回使用者端，而可 具有縮短檢索新聞的時間、降低使用者端的運算負擔、以 及減少資料傳輸頻寬等優點。 於一較佳實施例中，該股市新聞分類檢索系統更包括 有：一股名分類模組、以及一檢索模組。該股名分類模組 係用以自外界接收複數個股票名稱、以及各股票名稱所對 應之股票代碼與股票分類資料後，再將各股票名稱所對應 之股票代碼及股票分類資料加以群組成各股名分類，並據 以建立該股名分類資料庫。該檢索模組係用以接受來自外 界之一使用者端的一查詢請求，並依據該查詢請求中所包 含之至少一關鍵字所對應的股名分類，來從該新聞索引資 200832162 料庫中找出相對應於該股名分類的至少一新聞，並將該新 聞傳送給該使用者端。 為達上述目的’本發明之股市新聞分類檢索方法之— 實施例則係包括有下列步驟： 自外界之一證交所擷取複數個股票名稱、以及各股票 名稱所對應之一股票代碼與其所屬之一股票分類資料； 將各股票名稱所對應之股票代碼及股票分類資料來 加以群組成各股名分類，並據以建立該股名分類資料庫； 自外界之一新聞來源接收至少一新聞； 分析該新聞中所包含與股票相關之至少一關鍵字，該 關鍵字係包括至少下列其中之一:一股票名稱、一股票代 竭、及一股票分類名稱，並且，從該股名分_料庫;找 出與該關鍵字相對應之一股名分類； 建立該新聞與該股名分類之一索引關聯，並儲存於一 新聞索引資料庫中。 當新聞索引資料庫建域，當本發明之股市新聞分類 檢索方法更可包括下列之步驟： 接文來自外界之-使用者端的一查詢請求，於該查詢 请求中係包含與股票相關之至少一關鍵字； 從新聞索引資料庫中檢索出與該關鍵字相關聯之新 聞；以及 將符合檢索條件之新聞傳送給該使用者蠕。 200832162 【實施方式】 本發明之股市新聞分類檢索系統及方法的主要原 理，乃是在股市新聞分類檢索系統中增設股名分類資料庫 以及索引關聯模組。於股名分類資料庫中係儲存有依據各 股票名稱及其所對應之股票代碼與股票分類所構成之群 組化的股名分類資料。當系統自外界之一新聞來源接收到 新聞時，索引關聯模組將分析新聞中所包含之關鍵字，並 建立該新聞與該股名分類之索引關聯。之後，再將該新聞 與其所對應之索引關聯儲存於新聞索引資料庫中。因此， 當使用者端欲;^索某新聞時，僅需傳送特定關鍵字或代碼 至該系統’該系統便能自新聞索引資料庫中找出相符合之 新聞並傳送回使用者端，而可具有縮短檢索新聞的時間、 降低使用者端的運算負擔、以及減少資料傳輸頻寬等優 點。 請參酬二，其係林細之股市糊細檢索系統 之一較佳實施例的架構示意圖。基本上，本發明之股市新 聞分類檢索系統20係建立在由一無線通訊系統91 (例如 但不限於：GSM、CDMA、GPRS、PHS、WLAN 等）搭 配封閉網路92所構成之一電子交易環境中，並可供與^ 界之至少一手持式無線通訊裝置（亦即使用者端93)、一 新聞來源94、以及由金融證券服務業者（例如台灣證交 所95等冑）鍵立魏。於本雛實補+，該手持式 ·、、、線通訊裝置（使用者端93)係為一可執行以無線方式 下载新聞資料的智慧型手機（Smartph〇ne)、掌上型股票 200832162 機、或具行動通訊功能之個人數位助理（PDA)等等，其 藉由例如 GPRS (General Packet Radio Service，整合封包 無線電服務技術）等之行動通訊網路，來和行動通訊服務 業者所架設的無線通訊基地台96建立連線。此一行動通 訊服務業者再透過封閉網路92與本發明之股市新聞分類 檢索系統連線20，而使股市新聞分類檢索系統20與無線 通訊裝置（使用者端93)之間可以無線行動通訊的方式 建立連線並進行資料傳輸。 於本較佳實施例中，本發明之股市新聞分類檢索系統 20係包括有：一股名分類模組21、一股名分類資料庫22、 一索引關聯模組23、一新聞索引資料庫24、以及一檢索 模組25。該股名分類模組21係可自外界之證交所汜楱 收所有股票或期貨之股票名稱（例如「台積電」）、以及各 股票名稱所對應之股票代碼（例如「2330」）與股票分類 資料（例如「電子」、「機電」、「上市」、「集中Γ·.·等等）。 之後’再將各股票名稱所對應之股票代碼及股票分類資料 2以群組成各股名分類，並舰些股名分類（係由包括股 示名稱及紐應之股票代碼與所屬之股票分類等資料所 構成）儲存在該股名分類資料庫22甲。該索引關聯模組 23係可自外界之新聞來源94 (例如電子報業者等等）接 收若干新聞資料。當本發明之索引關聯模組23接收到新 聞時’並不會像習用技術般全部傳送給使用者端93 ,而 是會先分析各新聞令所包含與股票相關之至少一關鍵 字，並從該股名分類資料庫22 t找出與該綱字相對應 200832162 之一股名分類，並據以建立該新聞與該股名分類之一索引 關聯。然後，索引關聯模組23會把這些新聞連同其所對 應之該索引關聯一起儲存到新聞索引資料庫24中，以供 使用者93進行新聞檢索與下載之用。於本發明之一較佳 實施例中，該關鍵字係可包括至少下列其中之一：一股票 名稱、一股票代碼、及一股票分類名稱❹由於新聞資料中 引述有朋股票名稱與股票代碼的方式，大多是以例如『台 積電(2330)』的方式標示，因此，該索引關聯模組23可 藉由分析各新聞中所包含之特殊符號例如左刮號，，(，，或右 刮號等’來迅速地判別關鍵字尤其是股票代碼的存 在。藉此，當外界之使用者端93欲進行新聞檢索與下载 時’使用者端93只要發出一查詢請求給檢索模組25，其 中，該查詢請求中係包含由股票名稱、股票代碼或股票分 類三者至少其中之一所構成之關鍵字。之後，檢索模組 25將根據該關鍵字所對應的股名分類，來從該新聞索引 資料庫24中找出相對應於該股名分類的至少一新聞，並 將該新聞傳送給該使用者端93。並且，於另一較佳實施 例中，該檢索模組25更可接受來自該使用者端93的一建 立最愛明求’並依據該建立最愛請求中所包含之至少一關 鍵字所對應的股名分類，來儲存成對應於該使用者之一個 人化索引；之後，倘若新收到之新聞中，其所包含之關鍵 字所對應的股名分類係符合該個人化索引者，則將該新收 到之新聞主動推送給使用者端93。如此，由系統20傳送 給使用者端93的新聞都將是使用者感興趣的，不僅所需 12 200832162 傳輸之資料量大幅減少而可降低無線頻寬之佔用量、使用 者檢索新聞的時間也更為縮短、且使用者端％之運算負 載也更為減低者。 請參閱圖三所示，為本發明之股市新聞分類檢索方法 的一較佳實施例流程圖，其係包括有下列步驟： 步驟31 :自外界之證交所接收所有股票資料，包括 但不侷服於：複數個股票名稱、以及各股票名稱所對應之 _ 一股票代碼與其所屬之一股票分類資料。 步驟32 :依據該些股票資料來建立一股名分類資料 庫。將步驟31中所接收之各股票名稱所對應之股票代碼 及股票分類資料來加以群組成各股名分類，並據以建立該 股名分類資料庫並加以儲存。 步驟33·自外界之新聞來源接收至少一新聞。 步驟34 :分析新聞並建立其與股名分類之間的索引 關聯。其中，本步驟34又可細分為三個次步驟，包括： y 步驟341 :分析該新聞_所包含與股票相關之至少一關鍵 字；步驟342 :從該股名分類資料庫中找出與該關鍵字相 對應之知名分類，以及，步驟343 :建立該新聞與該股 名/7類之、索引關聯。於本步驟34中，該關鍵字係包括至 少下列其中之一：一股票名稱、一股票代碼、及一股票分 類名稱。而於本較佳實施例中，所述分析該新聞中所包含 之關鍵字的方式，乃是利用找尋該新聞中所包含之特殊符 號（例如到號），來擷取一股票代碼或是股票名稱以作為 該關鍵字。 13 200832162 步驟35 ··將新聞及其對應之索引關聯儲存於一新聞 索引資料庫中。 請參閲圖四，為本發明之使用者端欲自股市新聞分類 檢索系統檢索或下載新聞時之一較佳實施例流程圖。如圖 四所示，使用者端擷取新聞的方式可分成兩種，一種是即 時檢索與查詢，另一種則是先建立使用者最愛之選擇組合 並错存於股市新聞分類檢索系統中，以決定有哪些新聞會 被推送到使用者端。 如步驟41 ,當進行即時檢索與查詢時，首先使用者 端先發出一即時查詢請求至股市新聞分類檢索系統的檢 索模組’其中’於該查詢請求中係包含與股票相關之至少 一關鍵字❶之後，如步驟42 ,檢索模組會根據關鍵字所 對應之索引關聯，而從新聞索引資料庫中檢索出與該關鍵 字相關聯之新聞。最後，如步驟43，檢索模組會將符合 該查詢請求之條件的新聞回傳給外界之該使用者端。 如步驟44所示，當進行建立使用者最愛時，首先使 用者端先發出一建立最愛請求至股市新聞分類檢索系統 的檢索模組，其中，於該建立最愛請求中係包含與股票相 關之至少一關鍵字。之後，如步驟45，檢索模組從該股 名分類資料庫中找出與該至少一關鍵字相對應之至少一 股名分類，並將該至少一股名分類儲存成對應於該使用者 之一個人化索引。之後，如步驟46 ,當本發明之股市新 聞分類檢索系統自外界之新聞來源接收到新的新聞資料 時，將會檢查該新收到之新聞令，其所包含之關鍵字所對 200832162 應之股名分類疋否有符合該個人化索引者，若「是」則將 該新收到之新聞主動推送給使用者。 唯以上所述之實施例不應用於限制本發明之可應用 範圍，本發明之保護範圍應以本發明之申請專利範圍内容 所界定技術精神及其均等變化所含括之範園為主者。即大 凡依本發明申請專利範圍所做之均等變化及修飾，仍將不 失本發明之要義所在，亦不脫離本發明之精神和範圍，故 都應視為本發明的進一步實施狀況。 【圖式簡單說明】 圖一係為習知股市新聞檢索系統的示意圖。 圖一係為本發明之股市新聞分類檢索系統之一較佳實施 例的架構示意圖。 圖二係為本發明之股市新聞分類檢索方法的一較佳實施 例流程圖。 圖四係為本發明之使用者端欲自股市新聞分類檢索系統 檢索或下載新聞時之一較佳實施例流程圖。 【主要元件符號說明】 10〜習知的股市新聞檢索系統 11〜新聞來源 101〜新聞資料庫 12〜使用者端 121〜檢索模組 20〜本發明之股市新聞分類檢索系統 21〜股名分類模組 22〜股名分類資料庫 23〜索引關聯模組 24〜新聞索引資料庫 200832162 25〜檢索模組 31〜35、341〜343、40〜46〜流程步驟 91〜無線通訊系統 92〜封閉網路 93〜使用者端 94〜新聞來源 95〜證交所 96〜基地台200832162 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a stock market news system, method and method, and in particular, to shorten the inspection by establishing an index association between news and stock names. The time of the news, the reduction of the computing burden on the user side, and the reduction of the data transmission frequency S: - the city news classification search system and method test. [Prior Art] With the rapid advancement of information technology, people are experiencing explosive growth in the demand for instant and convenient information acquisition and communication. The portable wireless communication devices that are diversified in this way are constantly being developed in response to the convenience and needs of people's lives. For example, a variety of mobile phones (or Cellular Phone), smart phones (Smartph〇ne), tickets and personal mobile assistants (Personal faigitalAssistant (PDA)). In addition to the hard-wired devices of various wireless communication devices, many softwares and functions that can be implemented and used with the aforementioned hardware are continuously developed to make the user more convenient, more instantaneous, and more convenient. Manage money, work, play or grab information and more. For example, electronic trading of securities such as stocks or futures through wireless communication devices is one example. In addition to the functions of electronic ordering, the wireless communication devices used for electronic transactions of securities such as stocks or futures are an indispensable function for the retrieval and download of instant news. It can help users to more accurately determine the fluctuation trend of these securities 200832162, and to grasp the best trading opportunities. Please refer to Figure 1, which is a schematic diagram of a conventional stock market news retrieval system. At present, the conventional stock market news retrieval system 10, after receiving the news from the news source=end, stores the news in the news database 1〇1 on the one hand, and pushes the news in one direction on the other hand ( Push) to the wireless communication device (ie, user terminal 12) connected to it, without considering the actual needs of the user. These news materials, whether or not they are of interest to the user, are stored in the user terminal 12. Users must then search for and filter out the news they want to read, not only unable to provide the news that the user really needs, but also waste the limited transmission bandwidth of the bee-free communication. The memory storage capacity of the user terminal 12. Moreover, in the existing news search technology, most of the search modules 121 are disposed in the user terminal 12 (wireless communication device), and the time required for the user to perform the news search and the user end are increased. 12 computing burden. SUMMARY OF THE INVENTION The main object of the present invention is to provide a stock market news classification retrieval system and method, which shortens the time for retrieving news by establishing an index association between news and stock names and setting a retrieval module on the system side. The invention further reduces the computing burden of the user end and reduces the occupation of the data transmission bandwidth. Another object of the present invention is to provide a stock market news classification retrieval system and method, which can filter out the user true 200832162 according to the request of the user end. News of interest, and only those news that are of interest to these users will be transmitted to the user to save the bandwidth of the wireless communication and the usage of the memory of the user. To achieve the above objective, the stock market news classification and retrieval system provided by the present invention includes a stock classification database, an index correlation module and a news index database. There are multiple stock share categories stored in the stock classification database. Each share classification further includes a stock name, a stock code, and a stock classification. The index association module is configured to analyze the stock-related keywords included in the news, and find a stock name corresponding to the keyword from the stock classification database, and establish the news accordingly. Index association of stock name classification. The news is then stored in the news index database in association with its corresponding index. When the user wants to retrieve a certain news, only need to transmit a specific keyword or code to the system, the system can find the matching news from the news index database and transmit it back to the user end, but can have a shortened search. The time of news, the burden of computing on the user side, and the reduction of data transmission bandwidth. In a preferred embodiment, the stock market news classification system further includes: a name classification module and a search module. The stock name classification module is used to receive a plurality of stock names and the stock code and stock classification data corresponding to each stock name from the outside, and then group the stock code and stock classification data corresponding to each stock name. Each share is classified and the stock classification database is established accordingly. The search module is configured to accept a query request from a client of the outside world, and according to the stock name classification corresponding to at least one keyword included in the query request, to search from the news index resource 200832162 At least one news corresponding to the share name is sent and the news is transmitted to the user. In order to achieve the above object, the present invention relates to a stock market news classification search method. The embodiment includes the following steps: extracting a plurality of stock names from one of the external stock exchanges, and one stock code corresponding to each stock name and its One stock classification data; the stock code and stock classification data corresponding to each stock name are grouped into groups to be classified, and the stock classification database is established accordingly; at least one news is received from one of the external news sources. Analysing at least one keyword related to the stock included in the news, the keyword includes at least one of the following: a stock name, a stock exhaustion, and a stock classification name, and, from the stock name _ Repository; find one of the stock name categories corresponding to the keyword; establish the news to be indexed with one of the stock name categories, and store it in a news index database. When the news index database is built, the stock market news classification method of the present invention may further comprise the following steps: receiving an inquiry request from the outside world at the user end, wherein the query request includes at least one related to the stock. a keyword; retrieves news associated with the keyword from a news index database; and transmits news that meets the search criteria to the user. 200832162 [Embodiment] The main principle of the stock market news classification retrieval system and method of the present invention is to add a stock classification database and an index association module in the stock market news classification retrieval system. In the stock classification database, stock classification information based on the stock name and its corresponding stock code and stock classification is stored. When the system receives news from one of the external news sources, the index association module analyzes the keywords contained in the news and establishes an association of the news with the index of the stock name classification. The news is then stored in the news index database in association with its corresponding index. Therefore, when the user wants to download a certain keyword or code to the system, the system can find the matching news from the news index database and transmit it back to the user. It can have the advantages of shortening the time for retrieving news, reducing the computational burden on the user side, and reducing the bandwidth of data transmission. Please refer to the second, which is a schematic diagram of the structure of a preferred embodiment of the stock market paste retrieval system. Basically, the stock market news classification retrieval system 20 of the present invention is based on an electronic transaction environment composed of a wireless communication system 91 (such as, but not limited to, GSM, CDMA, GPRS, PHS, WLAN, etc.) with a closed network 92. At least one hand-held wireless communication device (ie, user terminal 93), a news source 94, and a financial securities service provider (such as the Taiwan Stock Exchange 95, etc.) can be used to establish the Wei. In this case, the handheld communication device (user end 93) is a smart phone (Smartph〇ne), palm stock 200832162, which can download news materials wirelessly. Or a personal digital assistant (PDA) with mobile communication function, etc., and a mobile communication network established by a mobile communication service provider by a mobile communication network such as GPRS (General Packet Radio Service) Taiwan 96 establishes a connection. The mobile communication service provider then connects to the stock market news classification and retrieval system 20 of the present invention through the closed network 92, so that the stock market news classification and retrieval system 20 and the wireless communication device (user terminal 93) can communicate wirelessly. Ways to establish a connection and transfer data. In the preferred embodiment, the stock market news classification and retrieval system 20 of the present invention includes: a name classification module 21, a stock classification database 22, an index association module 23, and a news index database 24. And a retrieval module 25. The stock classification module 21 is a stock name for all stocks or futures (such as "TSMC") from the external stock exchange, and the stock code (such as "2330") and stock classification corresponding to each stock name. Information (such as "electronics", "electromechanical", "listing", "concentration Γ···, etc.). Then 'the stock code and stock classification data corresponding to each stock name are classified by group name The classification of the shares of the ship is stored in the stock classification database 22A, which consists of the stock name including the name of the stock and the stock code of the stock and the stock classification of the stock. The index association module 23 is self-contained. External news sources 94 (e.g., electronic newspapers, etc.) receive certain news materials. When the index association module 23 of the present invention receives the news, 'it will not be transmitted to the user terminal 93 as in the conventional technology, but will First analyzing at least one keyword related to the stock in each news order, and finding a stock name corresponding to the 200832162 from the stock classification database 22 t, and establishing the news accordingly One of the stock name categories is indexed. The index association module 23 then stores the news in association with the corresponding index to the news index database 24 for the user 93 to perform news retrieval and downloading. In a preferred embodiment of the present invention, the keyword system may include at least one of the following: a stock name, a stock code, and a stock classification name. The manner in which the stock name and the stock code are quoted in the news material. Most of them are marked in the manner of, for example, TSMC (2330). Therefore, the index association module 23 can analyze special symbols included in each news such as the left scratch number, (, or right scratch number, etc. To quickly determine the existence of the keyword, especially the stock code, thereby, when the external user terminal 93 wants to perform the news retrieval and downloading, the user terminal 93 only needs to issue a query request to the retrieval module 25, wherein the query The request contains a keyword consisting of at least one of a stock name, a stock code, or a stock classification. Thereafter, the retrieval module 25 will be based on the key. The stock name corresponding to the word is classified to find at least one news corresponding to the stock name from the news index database 24, and the news is transmitted to the user terminal 93. And, in another preferred In an embodiment, the search module 25 further accepts a build favorite request from the user end 93 and stores the corresponding name according to the stock name corresponding to at least one keyword included in the create favorite request. Personalizing the index for one of the users; then, if the newly received news, the keyword classification corresponding to the keyword included in the news is in accordance with the personalized indexer, the newly received news is actively pushed to the news The user terminal 93. In this way, the news transmitted by the system 20 to the user terminal 93 will be of interest to the user, and not only the amount of data transmitted by the 12 200832162 is greatly reduced, but also the occupation of the wireless bandwidth can be reduced. The time for retrieving news is also shortened, and the computing load of the user terminal is also reduced. Referring to FIG. 3, a flow chart of a preferred embodiment of the stock market news classification and retrieval method of the present invention includes the following steps: Step 31: Receive all stock information from the external stock exchange, including but not Subject to: a plurality of stock names, and a stock code corresponding to each stock name and one stock stock. Step 32: Establish a generic classification database based on the stock data. The stock code and stock classification data corresponding to each stock name received in step 31 are grouped into groups, and the stock classification database is established and stored. Step 33. Receive at least one news from an external news source. Step 34: Analyze the news and establish an index association with its share name classification. The step 34 may be further divided into three sub-steps, including: y step 341: analyzing the news_containing at least one keyword related to the stock; step 342: finding and identifying from the stock classification database The well-known classification corresponding to the keyword, and, in step 343, the news is associated with the index of the stock name/7. In this step 34, the keyword includes at least one of the following: a stock name, a stock symbol, and a stock classification name. In the preferred embodiment, the manner of analyzing the keywords included in the news is to use a special symbol (such as a number) included in the news to retrieve a stock code or stock. The name is used as the keyword. 13 200832162 Step 35 · Store the news and its corresponding index associations in a news index database. Please refer to FIG. 4, which is a flow chart of a preferred embodiment of the user terminal of the present invention for retrieving or downloading news from the stock market news classification retrieval system. As shown in Figure 4, the way the user can retrieve news can be divided into two types, one is instant retrieval and query, and the other is to first establish a user's favorite combination of choices and store them in the stock market news classification retrieval system. Decide which news will be pushed to the user. In step 41, when performing an instant retrieval and query, the user first sends an instant query request to the search module of the stock market news classification retrieval system, wherein the query request includes at least one keyword related to the stock. After that, in step 42, the retrieval module retrieves the news associated with the keyword from the news index database according to the index corresponding to the keyword. Finally, in step 43, the retrieval module returns the news that meets the conditions of the query request to the user of the outside world. As shown in step 44, when establishing a user favorite, first, the user first sends a search module for establishing a favorite request to the stock market news classification retrieval system, wherein the establishment of the favorite request includes at least the stock-related A keyword. Then, in step 45, the retrieval module finds at least one name classification corresponding to the at least one keyword from the stock classification database, and stores the at least one name classification into corresponding to the user. A humanized index. Then, as in step 46, when the stock market news classification and retrieval system of the present invention receives new news materials from external news sources, it will check the newly received news order, and the keywords it contains are for 200832162. If there is a matching index in the stock name, if it is "Yes", the newly received news will be actively pushed to the user. The above-mentioned embodiments are not intended to limit the scope of application of the present invention, and the scope of the present invention should be based on the technical spirit defined by the content of the patent application scope of the present invention and its variations. It is to be understood that the scope of the present invention is not limited by the spirit and scope of the invention, and should be considered as a further embodiment of the invention. [Simple description of the diagram] Figure 1 is a schematic diagram of a conventional stock market news retrieval system. Figure 1 is a schematic diagram of the architecture of a preferred embodiment of the stock market news classification retrieval system of the present invention. Figure 2 is a flow chart of a preferred embodiment of the stock market news classification retrieval method of the present invention. Figure 4 is a flow chart of a preferred embodiment of the user terminal of the present invention for retrieving or downloading news from the stock market news classification retrieval system. [Main component symbol description] 10~ conventional stock market news retrieval system 11~ news source 101~ news database 12~user end 121~ retrieval module 20~ stock market news classification retrieval system 21~ stock name classification module of the present invention Group 22~Share Name Classification Database 23~Index Association Module 24~News Index Database 200832162 25~Search Modules 31~35, 341~343, 40~46~ Process Step 91~Wireless Communication System 92~Closed Network 93~user end94~news source95~stock exchange 96~base station